A magnitude 5.6 earthquake struck New Zealand's South Island on September 2, waking more than 6,000 people across various zones. ^[1] A new study published in Nature Communications suggests that machine-learning techniques could detect subtle changes in continuous vibrations around volcanoes to identify signs of an impending eruption hours beforehand. ^[2] Christchurch, Whakaari/White Island's sudden eruption in December 2019 killed 22 people and severely injured 25 others, marking New Zealand's deadliest volcanic disaster in recent history. ^[3] GeoNet, the platform monitoring seismic activity in New Zealand, confirmed the details of the magnitude 5.6 tremor. ^[4] Initial evaluations reported no injuries or material damage following the earthquake. ^[5] The earthquake occurred at 18:57 local time on September 2, originating at a depth of five kilometers approximately 35 kilometers northeast of Milford Sound. ^[6] The researchers used a cost-loss model to compare economic disruption caused by precautionary action with losses avoided when an eruption was successfully forecast, suggesting precautionary action could reduce preventable losses by 30-90%. ^[7] The study reanalyzed the performance of a machine-learning eruption forecaster at five volcanoes in New Zealand, Japan, Chile, and Russia using years of seismic data. ^[8] When tested on data it had not seen during training, the forecaster anticipated four of five eruptions at Whakaari, with a trade-off of approximately 15 days each year when a warning would have been in place without an eruption occurring. ^[9] In 2014, Japan's Mt Ontake erupted with little warning, killing 63 people, many of whom had been hiking near the summit. ^[10]